wip

CUDA/cudapit.cu

@@ -11,7 +11,7 @@
 #include <cstdint>
 uint64_t blocks = 1024;
 uint64_t threads = 1024;
-uint64_t nloop = 1000000;
+uint64_t nloop = 10000000;
 uint64_t npts = blocks*threads;
 std::vector<double> results;
 __global__ void init(double *arr,uint64_t nloop,uint64_t npts,int index) {